Corticosteroid therapy in regressive autism: a retrospective study of effects on the Frequency Modulated Auditory Evoked Response (FMAER), language, and behavior

Abstract

Background: Up to a third of children with Autism Spectrum Disorder (ASD) manifest regressive autism (R-ASD).They show normal early development followed by loss of language and social skills. Absent evidence-based therapies, anecdotal evidence suggests improvement following use of corticosteroids. This study examined the effects of corticosteroids for R-ASD children upon the 4 Hz frequency modulated evoked response (FMAER) arising from language cortex of the superior temporal gyrus (STG) and upon EEG background activity, language, and behavior. An untreated clinical convenience sample of ASD children served as control sample.

Methods: Twenty steroid-treated R-ASD (STAR) and 24 not-treated ASD patients (NSA), aged 3 - 5 years, were retrospectively identified from a large database. All study participants had two sequential FMAER and EEG studies;Landau-Kleffner syndrome diagnosis was excluded. All subjects' records contained clinical receptive and expressive language ratings based upon a priori developed metrics. The STAR group additionally was scored behaviorally regarding symptom severity as based on the Diagnostic and Statistical Manual IV (DSM-IV) ASD criteria list. EEGs were visually scored for abnormalities. FMAER responses were assessed quantitatively by spectral analysis. Treated and untreated group means and standard deviations for the FMAER, EEG, language, and behavior, were compared by paired t-test and Fisher's exact tests.

Results: The STAR group showed a significant increase in the 4 Hz FMAER spectral response and a significant reduction in response distortion compared to the NSA group. Star group subjects' language ratings were significantly improved and more STAR than NSA group subjects showed significant language improvement. Most STAR group children showed significant behavioral improvement after treatment. STAR group language and behavior improvement was retained one year after treatment. Groups did not differ in terms of minor EEG abnormalities. Steroid treatment produced no lasting morbidity.

Conclusions: Steroid treatment was associated with a significantly increased FMAER response magnitude, reduction of FMAER response distortion, and improvement in language and behavior scores. This was not observed in the non-treated group. These pilot findings warrant a prospective randomized validation trial of steroid treatment for R-ASD utilizing FMAER, EEG, and standardized ASD, language and behavior measures, and a longer follow-up period.Please see related article http://www.biomedcentral.com/1741-7015/12/79.

Figure 1

FMAER and corresponding FFT, before and after steroids in regressive autism. 4 Hz FMAER waveform data are shown within schematic ovals in vertex view with nose up, and left side of scalp to image left. The corresponding power spectra are shown to the immediate right. The top waveform and FFT displays were obtained prior to steroid administration. The bottom, corresponding displays were obtained after steroid administration. The vertical arrow to the lower left of each image represents 10 μV and the horizontal arrow beneath represents one second waveform length. The labels adjacent to the FMAER waveforms correspond to the standard EEG electrode 10–10 naming convention. Twenty-four electrodes’ waveforms are shown. The FFT power spectral data horizontal axis covers the 0–30 Hz range. Note the near absent 4 Hz FMAER waveform response before and excellent 4 Hz waveform response after steroid administration. Note the spread of spectral power over many frequencies in the FFT display before (above) which represents a distorted response. This contrasts to the nearly perfect 4 Hz response after steroid treatment (below) which shows little spectral spread (little distortion). For the vertex view display, waveforms are shown overlying their standard ‘10-10’ locations. For the FFT graphs, channel order from top to bottom is: F3, F4, C3, C4, P3, P4, O1, O2, Fp1, Fp2, F7, F8, T7, T8, P7, P8, FT9, FT10, TP9, TP10, Fz, Cz, Pz, Oz. The common average reference is utilized for the displayed data (a reference free or ‘rfr’ technique) [10]. Abbreviations: A = anterior, P = posterior, L-left, R = right, FMAER = 4 Hz frequency modulated auditory evoked response, FFT = fast Fourier transform - power spectrum analysis shown as μV²/Hz, μV = microvolt, Hz = Hertz or cycles per second.